def post_process_normal(spectrum_in: SpectrumType, min_peaks: int = 10) \
-> Union[SpectrumType, None]:
"""Normal processing of spectra for Spec2Vec
Parameters
----------
spectrum_in:
Input spectrum.
min_peaks:
Minimum number of peaks to pass the spectrum (otherwise -> None)
"""
if spectrum_in is None:
return None
s = spectrum_in.clone()
s = normalize_intensities(s)
if any(np.isnan(s.peaks[1])):
return None # remove spectra that have all intensities 0
s = select_by_mz(s, mz_from=0, mz_to=1000)
s = require_minimum_number_of_peaks(s, n_required=min_peaks)
s = reduce_to_number_of_peaks(s, n_required=min_peaks, ratio_desired=0.5)
if s is None:
return None
# remove low peaks unless less than 10 peaks are left
s_remove_low_peaks = select_by_relative_intensity(s, intensity_from=0.001)
if len(s_remove_low_peaks.peaks) >= 10:
s = s_remove_low_peaks
# add losses to normally processed spectra
s = add_losses(s, loss_mz_from=5.0, loss_mz_to=200.0)
return s
def test_reduce_to_number_of_peaks_ratio_given_but_no_parent_mass():
"""A ratio_desired given without parent_mass should not result in changes."""
mz = numpy.array([10, 20, 30, 40], dtype="float")
intensities = numpy.array([0, 1, 10, 100], dtype="float")
spectrum_in = Spectrum(mz=mz, intensities=intensities)
spectrum = reduce_to_number_of_peaks(spectrum_in, n_required=4, ratio_desired=0.1)
assert spectrum == spectrum_in, "Expected the spectrum to remain unchanged."
def test_reduce_to_number_of_peaks_no_changes(metadata):
mz = numpy.array([10, 20, 30, 40], dtype="float")
intensities = numpy.array([0, 1, 10, 100], dtype="float")
spectrum_in = SpectrumBuilder().with_mz(mz).with_intensities(
intensities).with_metadata(metadata).build()
spectrum = reduce_to_number_of_peaks(spectrum_in)
assert spectrum == spectrum_in, "Expected no changes."
def test_reduce_to_number_of_peaks_no_params():
"""Use default parameters."""
mz = numpy.array([10, 20, 30, 40], dtype="float")
intensities = numpy.array([0, 1, 10, 100], dtype="float")
spectrum_in = Spectrum(mz=mz, intensities=intensities)
spectrum = reduce_to_number_of_peaks(spectrum_in)
assert spectrum == spectrum_in, "Expected no changes."
def post_process_md(spectrum_in: SpectrumType,
low_int_cutoff: float = 0.05,
min_peaks: int = 10,
max_peaks: int = 30) -> Union[SpectrumType, None]:
"""Processing of spectra that are used for mass difference extraction
Parameters
----------
spectrum_in:
Input spectrum.
low_int_cutoff:
Lower intensity cutoff for the peaks selected for MD
min_peaks:
Minimum number of peaks to pass the spectrum (otherwise -> None)
max_peaks:
Maximum number of peaks allowed in the spectrum (ranked on intensity)
"""
if spectrum_in is None:
return None
s = spectrum_in.clone()
# remove precurzor_mz from spectra so neutral losses don't end up in MDs
s = remove_precursor_mz_peak(s)
s = normalize_intensities(s)
if any(np.isnan(s.peaks[1])):
return None # remove spectra that have all intensities 0
s = select_by_mz(s, mz_from=0, mz_to=1000)
s = require_minimum_number_of_peaks(s, n_required=min_peaks)
s = reduce_to_number_of_peaks(s, n_required=min_peaks, ratio_desired=0.5)
if s is None:
return None
# remove low peaks unless less than 10 peaks are left
s_remove_low_peaks = select_by_relative_intensity(s, intensity_from=0.001)
if len(s_remove_low_peaks.peaks) >= 10:
s = s_remove_low_peaks
# do an additional removal step with a different intensity cutoff
s_second_peak_removal = select_by_relative_intensity(
s, intensity_from=low_int_cutoff)
if len(s_second_peak_removal.peaks) >= 10:
s = s_second_peak_removal
# reduce to top30 peaks
s = reduce_to_number_of_peaks(s, n_required=min_peaks, n_max=max_peaks)
return s
def test_reduce_to_number_of_peaks_n_max_4():
mz = numpy.array([10, 20, 30, 40, 50], dtype="float")
intensities = numpy.array([1, 1, 10, 20, 100], dtype="float")
spectrum_in = Spectrum(mz=mz, intensities=intensities)
spectrum = reduce_to_number_of_peaks(spectrum_in, n_max=4)
assert len(spectrum.peaks) == 4, "Expected that only 4 peaks remain."
assert spectrum.peaks.mz.tolist() == [20., 30., 40., 50.], "Expected different peaks to remain."
def test_reduce_to_number_of_peaks_desired_5_check_sorting():
"""Check if mz and intensities order is sorted correctly """
mz = numpy.array([10, 20, 30, 40, 50, 60], dtype="float")
intensities = numpy.array([5, 1, 4, 3, 100, 2], dtype="float")
spectrum_in = Spectrum(mz=mz, intensities=intensities)
spectrum = reduce_to_number_of_peaks(spectrum_in, n_max=5)
assert spectrum.peaks.intensities.tolist() == [5., 4., 3., 100., 2.], "Expected different intensities."
assert spectrum.peaks.mz.tolist() == [10., 30., 40., 50., 60.], "Expected different peaks to remain."
def apply_my_filters(s):
"""This is how a user would typically design his own pre- and post-
processing pipeline."""
s = default_filters(s)
s = add_parent_mass(s)
s = normalize_intensities(s)
s = reduce_to_number_of_peaks(s, n_required=10, ratio_desired=0.5)
s = select_by_mz(s, mz_from=0, mz_to=1000)
s = add_losses(s, loss_mz_from=10.0, loss_mz_to=200.0)
s = require_minimum_number_of_peaks(s, n_required=5)
return s
def spectrum_processing(s):
"""This is how one would typically design a desired pre- and post-
processing pipeline."""
s = default_filters(s)
s = add_precursor_mz(s)
s = normalize_intensities(s)
s = reduce_to_number_of_peaks(s, n_required=5, ratio_desired=0.5, n_max=500)
s = select_by_mz(s, mz_from=0, mz_to=1000)
s = add_losses(s, loss_mz_from=10.0, loss_mz_to=200.0)
s = require_minimum_number_of_peaks(s, n_required=5)
return s
def test_reduce_to_number_of_peaks_no_params_w_parent_mass():
"""Use default parameters with present parent mass."""
mz = numpy.array([10, 20, 30, 40], dtype="float")
intensities = numpy.array([0, 1, 10, 100], dtype="float")
spectrum_in = Spectrum(mz=mz,
intensities=intensities,
metadata={"parent_mass": 50})
spectrum = reduce_to_number_of_peaks(spectrum_in)
assert spectrum == spectrum_in, "Expected no changes."
def test_reduce_to_number_of_peaks_required_2_desired_3():
"""Here: ratio_desired * parent_mass is 3, more than n_required."""
mz = numpy.array([10, 20, 30, 40], dtype="float")
intensities = numpy.array([0, 1, 10, 100], dtype="float")
spectrum_in = Spectrum(mz=mz, intensities=intensities,
metadata={"parent_mass": 20})
spectrum = reduce_to_number_of_peaks(spectrum_in, n_required=3, n_max=4, ratio_desired=0.1)
assert len(spectrum.peaks) == 3, "Expected that only 3 peaks remain."
assert spectrum.peaks.mz.tolist() == [20., 30., 40.], "Expected different peaks to remain."
def test_reduce_to_number_of_peaks_ratio_given_but_no_parent_mass():
"""A ratio_desired given without parent_mass should raise an exception."""
mz = numpy.array([10, 20, 30, 40], dtype="float")
intensities = numpy.array([0, 1, 10, 100], dtype="float")
spectrum_in = Spectrum(mz=mz, intensities=intensities)
with pytest.raises(Exception) as msg:
_ = reduce_to_number_of_peaks(spectrum_in,
n_required=4,
ratio_desired=0.1)
expected_msg = "Cannot use ratio_desired for spectrum without parent_mass."
assert expected_msg in str(
msg.value), "Expected specific exception message."
def test_reduce_to_number_of_peaks(mz, intensities, metadata, params,
expected):
spectrum_in = SpectrumBuilder().with_mz(mz).with_intensities(
intensities).with_metadata(metadata).build()
n_required, n_max, ratio_desired = params
spectrum = reduce_to_number_of_peaks(spectrum_in,
n_required=n_required,
n_max=n_max,
ratio_desired=ratio_desired)
assert len(
spectrum.peaks) == len(expected), "Expected that only 4 peaks remain."
assert spectrum.peaks.mz.tolist(
) == expected, "Expected different peaks to remain."
def post_process(s):
s = normalize_intensities(s)
s = select_by_mz(s, mz_from=0, mz_to=1000)
s = require_minimum_number_of_peaks(s, n_required=10)
try:
s = reduce_to_number_of_peaks(s, n_required=10, ratio_desired=0.5)
except:
pass
if s is None:
return None
s_remove_low_peaks = select_by_relative_intensity(s, intensity_from=0.001)
if len(s_remove_low_peaks.peaks) >= 10:
s = s_remove_low_peaks
s = add_losses(s, loss_mz_from=5.0, loss_mz_to=200.0)
return s
def test_reduce_to_number_of_peaks_desired_5_check_sorting():
"""Check if mz and intensities order is sorted correctly """
mz = numpy.array([10, 20, 30, 40, 50, 60], dtype="float")
intensities = numpy.array([5, 1, 4, 3, 100, 2], dtype="float")
metadata = {"parent_mass": 20}
spectrum_in = SpectrumBuilder().with_mz(mz).with_intensities(
intensities).with_metadata(metadata).build()
spectrum = reduce_to_number_of_peaks(spectrum_in, n_max=5)
assert spectrum.peaks.intensities.tolist() == [
5., 4., 3., 100., 2.
], "Expected different intensities."
assert spectrum.peaks.mz.tolist() == [
10., 30., 40., 50., 60.
], "Expected different peaks to remain."
def test_reduce_to_number_of_peaks_n_max_4():
"""Test setting n_max parameter."""
mz = numpy.array([10, 20, 30, 40, 50], dtype="float")
intensities = numpy.array([1, 1, 10, 20, 100], dtype="float")
spectrum_in = SpectrumBuilder().with_mz(mz).with_intensities(
intensities).build()
spectrum = reduce_to_number_of_peaks(spectrum_in, n_max=4)
expected = numpy.array([20, 30, 40, 50], dtype="float")
assert len(
spectrum.peaks) == len(expected), "Expected that only 4 peaks remain."
numpy.testing.assert_array_equal(
spectrum.peaks.mz,
expected,
err_msg="Expected different peaks to remain.")
def test_reduce_to_number_of_peaks_set_to_none():
"""Test is spectrum is set to None if not enough peaks."""
set_matchms_logger_level("INFO")
mz = numpy.array([10, 20], dtype="float")
intensities = numpy.array([0.5, 1], dtype="float")
spectrum_in = SpectrumBuilder().with_mz(mz).with_intensities(
intensities).with_metadata({
"parent_mass": 50
}).build()
with LogCapture() as log:
spectrum = reduce_to_number_of_peaks(spectrum_in, n_required=5)
assert spectrum is None, "Expected spectrum to be set to None."
log.check(
('matchms', 'INFO', "Spectrum with 2 (<5) peaks was set to None."))
reset_matchms_logger()
def spectrum_processing_s2v(
spectrum: SpectrumType,
**settings: Union[int, float]) -> Union[SpectrumType]:
"""Spectrum processing required for computing Spec2Vec scores.
Args:
----------
spectrum:
Spectrum to process
mz_from:
Peaks below this value are removed. Default = 10.0
mz_to:
Peaks above this value are removed. Default = 1000.0
n_required
Number of minimal required peaks for a spectrum to be considered.
n_max
Maximum number of peaks to be kept per spectrum. Default is 1000.
loss_mz_from
Minimum allowed m/z value for losses. Default is 0.0.
loss_mz_to
Maximum allowed m/z value for losses. Default is 1000.0.
"""
settings = set_spec2vec_defaults(**settings)
spectrum = select_by_mz(spectrum,
mz_from=settings["mz_from"],
mz_to=settings["mz_to"])
spectrum = reduce_to_number_of_peaks(spectrum,
n_required=settings["n_required"],
n_max=settings["n_max"])
spectrum = add_losses(spectrum,
loss_mz_from=settings["loss_mz_from"],
loss_mz_to=settings["loss_mz_to"])
assert spectrum is not None, \
"Expects Spectrum that has high enough quality and is not None"
return spectrum
def test_empty_spectrum():
spectrum_in = None
spectrum = reduce_to_number_of_peaks(spectrum_in)
assert spectrum is None, "Expected different handling of None spectrum."
